January 2019
Intermediate to advanced
606 pages
15h 1m
English
Sometimes, slow performance may not be due to an underlying fault; it may just be that the number and the type of client requests may have exceeded the capability of the hardware. Whether this is due to a number of separate workloads all running at the same time, or just a slow general increase over a period of time, if you are capturing the number of client requests across your cluster, this should be easy to trend. If the increased workload looks like it's permanent, then the only solution is to add some additional hardware.
Read now
Unlock full access